Metaphors make the world ⊗ The OpenAI endgame ⊗ Apple Vision Pro urban encounter

No.298 — Overpromising & stumbling Bambis ⊗ The scarcity of the long-term ⊗ Are fictional dystopias blocking us from better futures? ⊗ Measuring AI’s environmental impacts

Metaphors make the world ⊗ The OpenAI endgame ⊗ Apple Vision Pro urban encounter
Metaphors make the world. Made with Midjourney.

Metaphors make the world

Fascinating piece at Aeon on how metaphors play a significant role in language and thought. They go beyond casual speech and provide the foundation for how we conceptualize abstract concepts and much of the world. They can influence the opinions and decisions of individuals in a variety of settings (health, love, elections, pandemics), and also scaffold science—helping to formulate and frame theories.

The author explains that metaphors run so deep and have such an impact on our understanding of certain ideas, like “battling cancer,” that they can greatly skew how we view certain situations and act upon them, sometimes to our detriment and society’s. He argues that their influence is such that we should consider making efforts to replace some metaphors by new ones that might better serve our goals. (Via keels on RADAR.)

Given the nature of our metaphorical minds, it is worth asking: are our conceptual metaphors apt? We owe it to ourselves and others to reflect on the appropriateness of the metaphors we employ to frame the world. These choices – conscious or not – can be constructive or disastrous. […]

The particularly frightening power of conceptual metaphors is not that a group is seen unfavourably and then, to emphasise this point of view, referred to by dehumanising metaphors. Rather, it is that the metaphorical construction used to frame a particular group in the first place is a reason why the other group sees them that way. […]

Metaphors are (metaphorically) woven into the fabric of our language and thought, shaping how we grasp and articulate abstract concepts. We should therefore feel free to prudently explore alternative metaphors and judge whether they perform better. A collective effort to notice and change the metaphors we use has enormous potential to reduce individual and societal harm.

The OpenAI endgame

Mike Loukides considers the New York Times’ copyright infringement lawsuit against OpenAI and proposes that the latter basically can’t lose. Whatever the exact outcome, they’ll come out of it with a de-facto price on training data that could deter competitors. This could leave generative AI in the hands of a small number of unassailable monopolies, limiting AI development to a few wealthy players who make private agreements with publishers.

Loukides also explores the potential impact on model developers, academic research, open source models, and the future of AI development, while emphasizing the importance of transparency and the potential risks of limiting AI development to a few wealthy players.

This is chilling: if all AI applications go through one of a small group of monopolists, can we trust those monopolists to deal honestly with issues of bias? AI developers have said a lot about “alignment,” but discussions of alignment always seem to sidestep more immediate issues like race and gender-based bias. […]

Without data transparency, how will it be possible to understand biases that are built in to any model? Understanding those biases will be important to addressing the harms that models are doing now, not hypothetical harms that might arise from sci-fi superintelligence. […]

What will AI be in the future? Will there be a proliferation of models? Will AI users, both corporate and individuals, be able to build tools that serve them? Or will we be stuck with a small number of AI models running in the cloud and being billed by the transaction, where we never really understand what the model is doing or what its capabilities are? That’s what the endgame to the legal battle between OpenAI and the Times is all about.

Can Apple Vision Pro survive an urban encounter?

In this piece from Bryan Boyer’s newsletter on urban technology, he considers AR from an uncommon angle; the urban planning process. Using a number of screenshots, Boyer compares the domestic focus of Apple’s promotional material to the urban focus of Google Glass back in the day, and speculates that Apple may be downplaying the device’s potential for use in urban environments due to current technical and social constraints.

He gives the Swiss baugespann (“story poles” forming the ghost of proposed buildings) as a useful example of “AR.” Boyer is “bullish on augmented reality in cities for a very specific use: enabling communities to preview and debate proposed developments in their city. In other words, seeing the city anew and thus empowering residents to have a stronger voice in how that environment is shaped by development.”

By contrast to Google Glass’s positioning as a device to help you navigate the city, Apple Vision Pro is advertised explicitly as a “spatial computer” to (implicitly) help you navigate inwards: to your work, to mediated fantasy worlds, to your memories via unique 3d videos, and to yourself via a meditation app. […]

Social norms relating to the use of technology in public change at the pace of society, so I read Apple's emphasis on indoor environments as a shrewd play to get more people comfortable with these devices in private. […]

The more tools available to help communities make future scenarios rapidly and affordably tangible, the better. The more tools available to let communities see proposed urban developments from their own point of view, the better. The more diversity and exploration in the design and conceptualization of these tools and processes, the better.


§ Overpromising & stumbling Bambis. “In parallel, I’d like to see ‘new things’ launched with significantly more openness and honesty to their flaws. We need to get beyond the notion that uncertainty is a weakness. Let’s find a way to present new ideas as a representation of the best we can do right now, and that we’re open to change. Any ‘new thing’ won’t have got it right the first time, and there should always room for pivots, evolutions and changes of direction.”


§ The scarcity of the long-term. “If let alone, an advanced civilization, could over many millennia, invent AIs and other tools that would allow it to invent almost any material it could imagine. There would be no resource in the universe it could not synthesize at home. In that sense, there would be no material and energy scarcity. Perhaps no knowledge scarcity either. The only real scarcity would be of a long attention span. That is not something you can buy, or download. You’d need some new tools for transmitting values and missions into the future.”

Futures, Fictions & Fabulations

Are fictional dystopias blocking us from better futures?
“Since the 1970s the prevailing vision of the future in popular culture has tended towards the dystopian. Commentators of all stripes — from celebrated movie critics to novelists and today’s ‘effective accelerationists’ — have addressed the lack of blue-sky thinking.”

The META Trending Trends: 2024
“META Trends are a distillation of all of the year’s reported trends — an annual meta-analysis answering, ‘What are the most-mentioned trends across 70+ global cultural forecasts?’”

How science fiction helps us harness the power of imagination
Excellent episode of Nora Young’s Spark podcast. “From inspiring innovation, to framing collective visions of the future — a look at the imaginative power of science fiction.”

Algorithms, Automation, Augmentation

OpenAI’s Sora
“An AI model that can create realistic and imaginative scenes from text instructions.” It’s been made available to red teamers for testing, not yet publicly available. If this is real output, like they say it is, it’s absolutely bonkers.

China turns to AI tablets for students after tutoring crackdown
“AI educational tablets are gaining popularity in China, with a push from tech companies like iFlytek and Baidu. The devices are popular in smaller, less-developed cities, where after-school education is harder to come by.”

Measuring AI’s environmental impacts requires empirical research and standards
“The [Artificial Intelligence Environmental Impacts Act of 2024] calls for the Environmental Protection Authority to direct a study on the environmental impacts of artificial intelligence (AI) across its lifecycle, from the mining of rare earth minerals and the manufacturing and disposal of chips, to the training and use of models.”

Asides

Your Futures Thinking Observatory